Roof Temperature Histories in Matched Attics in ... · roof temperature histories presented in this report provide reasonable estimates of actual thermal loads. In addition, the histories

2 Roof Temperature Histories in Matched Attics in Mississippi and Wisconsin Jerrold E. Winandy H. Michael Barnes Cherilyn A. Hatfield

United States Department of Agriculture Forest Service Forest Products Laboratory Research Paper FPL−RP−589

Abstract To address the problem of defining actual field temperatures of various wood components in wood-framed roof systems, roof temperatures were monitored in test structures situated in the northern and southern United States (Madison, Wis-consin, and Starkville, Mississippi, respectively). The field exposure structures were intended to simulate the attics of multifamily wood-framed structures for which Model Build-ing Codes sometimes allow the use of fire-retardant-treated roof sheathing. The structures were instrumented to monitor interior attic air, exterior air, inner and outer plywood roof sheathing, and internal rafter temperatures in dry white-shingled structures and both dry and heavily humidified black-shingled structures. Temperatures were recorded from January 1992 through December 1999 in Wisconsin and from January 1996 through December 1999 in Mississippi. The Mississippi exposure generally induced 5°C to 10°C higher temperatures than did the Madison exposure, though the difference in annual maximum “1-h average” temperature of both exposures was usually no more than 3°C to 4°C.

December 2000 Winandy, Jerrold E.; Barnes, H. Michael; Hatfield, Cherilyn A. 2000. Roof

temperature histories in matched attics in Mississippi and Wisconsin. Res. Pap. FPL–RP–589. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. 24 p.

A limited number of free copies of this publication are available to the public from the Forest Products Laboratory, One Gifford Pinchot Drive, Madison, WI 53705–2398. Laboratory publications are sent to hundreds of libraries in the United States and elsewhere.

The Forest Products Laboratory is maintained in cooperation with the University of Wisconsin.

The use of trade or firm names is for information only and does not imply endorsement by the U.S. Department of Agriculture of any product or service.

The United States Department of Agriculture (USDA) prohibits discrimina-tion in all its programs and activities on the basis of race, color, national origin, sex, religion, age, disability, political beliefs, sexual orientation, or marital or familial status. (Not all prohibited bases apply to all programs.) Persons with disabilities who require alternative means for communication of program information (Braille, large print, audiotape, etc.) should contact the USDA’s TARGET Center at (202) 720–2600 (voice and TDD). To file a complaint of discrimination, write USDA, Director, Office of Civil Rights, Room 326-W, Whitten Building, 1400 Independence Avenue, SW, Wash-ington, DC 20250–9410, or call (202) 720–5964 (voice and TDD). USDA is an equal opportunity provider and employer.

Overall, black-shingled roof systems tended to be 5°C to 10°C warmer during the midafternoon of a sunny day than were comparable white-shingled roof systems. Few differ-ences in plywood roof sheathing temperatures were noted between dry and heavily humidified structures. Attic air-space and rafter temperatures were generally 5°C to 10°C cooler in humidified attics than in dry attics.

The major difference in the temperature of wood components in the Wisconsin and Mississippi structures occurred during the winter, when temperatures were as much as 20°C lower in Wisconsin.

Keywords: roof temperature, plywood, roof sheathing, rafter, thermal degrade, fire-retardant treatment, shingles, attic

Contents Page

Problem............................................................................... 1

Objectives ........................................................................... 1

Background......................................................................... 1

Methods .............................................................................. 2

Exposure Structures ........................................................ 2

Recording of Temperatures............................................. 3

Measurement of Moisture Content.................................. 5

Results and Discussion........................................................ 5

Sheathing Temperatures.................................................. 5

Exceedance Temperatures .............................................. 5

Maximum Temperatures ................................................. 5

Annual Temperature Trends ........................................... 6

Peak Temperature Trends ............................................... 9

Moisture Content .......................................................... 11

Concluding Remarks......................................................... 11

Acknowledgments............................................................. 16

References......................................................................... 16

Appendix A—Exceedance Temperature Data for WI Structures .......................................................................... 17

Appendix B—Exceedance Temperature Data for MS Structures .......................................................................... 22

Roof Temperature Histories in Matched Attics in Misissippi and Wisconsin Jerrold E. Winandy, Research Wood Scientist Forest Products Laboratory, Madison, Wisconsin

H. Michael Barnes , Professor Mississippi State University, Mississippi

Cherilyn, A. Hatfield, Statistician Forest Products Laboratory, Madison, Wisconsin

Problem In the late 1980s, the degradation of wood treated with some fire retardant (FR) chemicals in roof systems became a prob-lem of major national significance. Thousands of cases were reported throughout the eastern United States (NAHB 1990). Our understanding of this deterioration in serviceability caused by thermal degrade has been limited because we have been unable to specifically correlate laboratory experiments using steady-state and cyclic temperature exposures with actual diurnal field temperature histories of FR-treated roof sheathing plywood. This lack of correlation has inhibited our ability to predict thermal-induced degradation of FR-treated plywood in the field from thermal degradation rates derived in the laboratory.

In many reports of field problems with FR-treated plywood roof sheathing, improper or nonfunctional ventilation of roof systems was attributed as a co-contributor to thermal degra-dation (LeVan and Collet 1989, NAHB 1990). This has resulted in questions about the relationship between roof system temperatures in damp, nonventilated attics compared with normally dry systems and the potential for synergy between roof temperatures and roof moisture.

Objectives The primary objective of this work is to collect actual field data documenting the actual thermal load history of various wood components in traditional light-framed structures. The roof temperature histories presented in this report provide reasonable estimates of actual thermal loads. In addition, the histories provide insight into the effects of shingle color, attic humidity, and climate on thermal loads and a data base for modeling the residual serviceability of roof sheathing for known wood thermal degradation rates (Lebow and Winandy 1999). The secondary objective of this work is to use matched laboratory and field data on mechanical properties to develop specific laboratory-to-field correlations. These correlations will provide specific guidance for deriving the

cyclic factor used in Section 7 of ASTM D6305 (ASTM 1999) because the current cyclic factor is an arbitrarily derived estimate rather than a calculated estimate. This report specifically addresses the primary objective and presents the findings from one of more than a dozen interrelated studies conducted in a 10-year research program to develop residual serviceability models for roof sheathing (Winandy 2000).

Background Heyer (1963) reported temperature histories for wall and roof systems for six houses and one office building for 1 week to 2 consecutive summers (June–August). The houses were located in Tucson, Arizona; Athens, Georgia; Portland, Oregon; Diboll, Texas; and Madison, Wisconsin. The office building, which had served as the original headquarters of the Forest Products Society, was located in Madison, Wisconsin. In any one year, maximum roof temperatures were found to reach as high as 75°C but the cumulative duration of tem-peratures over 70°C did not exceed 21 h, and the cumulative duration of temperatures over 65°C did not exceed 64 h. This was thought to be important considering that design stan-dards for wood (AF&PA 1997) require a strength property adjustment for sustained exposures above 37.8°C and greater adjustment for prolonged exposures up to 65.6°C.

Roof temperatures attained by structures have been modeled. Ozkan (1993) and Wilkes (1989) reported on surface and various component temperatures in flat roof systems. Wood sheathing temperatures were not considered. In the study by Ozkan (1993), which was conducted in a very hot and dry area in Arabia, temperatures of roofing surfaces of a field station reached 93°C during a 1½-year period (April 1989 to November 1990). The primary use of the station was to observe the effects of weathering and to measure the tem-peratures of the bituminous and polymeric waterproofing membranes and thermal insulation materials. In the study by Wilkes (1989), temperatures of metal roofs in eastern Ten-nessee reached as high as 73°C during January and May. For more exposure temperature histories for shingles, the reader

2

is referred to studies by the National Bureau of Standards (NBS 1979) and Blackenstowe (1987). The temperature histories discussed hereafter pertain to wood components of roof systems.

Computer models have been developed that predict the aver-age temperature and moisture content of plywood roof sheathing and other lumber roof members based on various construction details, materials, ventilation factors, and solar gain (radiation load) for the roof (APA 1989, ASTM 1988, TenWolde 1997, Wilkes 1989). The APA–Engineered Wood Association reported on a model that predicts temperatures of plywood roof sheathing under a black membrane in flat-roofed systems. This model predicts that sheathing tempera-tures of 65°C, 70°C, 75°C, and 80°C might be exceeded for up to 36, 13, 5, and 2 h, respectively, over the course of an average year in Hartford, Connecticut (APA 1989). Wilkes (1989) developed and reported a predictive roof temperature model for multi-layer nonwood roof systems. This model does not account for moisture flux, which may be critical in wood roof systems.

TenWolde developed and later verified (TenWolde 1988 and 1997, respectively) a predictive roof temperature model. In the 1997 report, he described a predictive roof temperature model especially for sloped wood-based roof systems. This model shows that the surface temperature of plywood roof sheathing is dominated by solar gain and the heat exchange between the surface and ambient air, not by attic ventilation. Diurnal (daily cyclic) temperature variation and hourly sheathing temperature histories are also influenced by the radiant energy absorptivity of the roofing surface, roof pitch, and, to a lesser extent, insulation and attic ventilation. The TenWolde model predicts that wet plywood sheathing dries quickly under warm summer conditions, even if ventilation is minimal. For example, if plywood is installed at 60% mois-ture content, the moisture content is roughly 15% after 1 week and falls to 8% in roughly 2 weeks. The model also indicates that the absorbtivity of solar (radiant) energy by the roofing material has the greatest effect on increasing or reducing the average temperature of the plywood roof sheath-ing. If the absorptivity of the roofing material is 0.92, the model predicts the maximum hourly temperature for the roof sheathing plywood as 60°C and the maximum predicted exterior roof membrane temperature as 66°C. If the absorp-tivity is changed to 0.2, supposedly representing a metal roof system, both the maximum predicted sheathing temperature and maximum predicted membrane temperature drop to 35°C. Roof pitch has only a moderate influence on reducing the exterior surface temperature and the average temperature of the plywood. The model also predicts that the presence of insulation installed directly on the underside of the sheathing has virtually no influence on sheathing temperature on the top surface, but raises the average sheathing temperature relative to that of the top surface. When the ventilation rate in uninsulated systems is increased from 8 to about 21 air

changes per hour, almost no decrease of the top surface sheathing temperature or the average sheathing temperature is predicted.

In 1992, a test facility was constructed at the Building Re-search Council of the University of Illinois to measure heat transfer, moisture movement, and airflow in typical residen-tial attic structures under natural conditions (Rose 1992). The results of that study showed that during the summer in central Illinois, attic ventilation could lower attic air temperature by 15.5°C but sheathing temperature by only 5°C; information on rafter temperatures was not given. Attic ventilation had only a minor effect on roof shingle temperature. The findings of Rose (1995) affirmed the earlier predictions of TenWolde (1988) and were then used to verify the TenWolde model (1997). Eventually, the TenWolde model was used to predict roof temperature histories for plywood roof sheathing at a dozen locations across the Unites States. Those predictions were used to predict engineering design adjustments for FR-treated plywood roof sheathing in ASTM Standard D6305–98 (ASTM 1999).

Methods Exposure Structures In the summer of 1991, five field exposure structures were constructed near Madison, Wisconsin (43° latitude). In Madison, the average incidence angle of sunlight is 19.5° from the southern horizon on the winter solstice (December 21) and 43° on the summer solstice (June 21). The annual average declination angle is 31.25°. The Wisconsin exposure structures (WI structures) were constructed to face south in a shadeless area open to direct sunlight. The structures were spaced far enough apart to prevent any one structure from shading the next structure. The construction of the WI struc-tures was fully described by Winandy and Beaumont (1995).

In 1994, a USDA Competitive Grant was received to con-struct and monitor matched exposure structures at the Missis-sippi Forest Products Laboratory, Mississippi State Univer-sity, in Starkville, Mississippi (33.5° latitude). This research was part of an ongoing effort to relate temperatures in matched northern to southern U.S. roof systems (Barnes and others 1993). In Starkville, the average incidence angle of sunlight is 32.3° from the southern horizon on the winter solstice and 74.8° on the summer solstice. The annual aver-age declination angle is 53.5°. The five exposure structures in Mississippi (MS structures) were constructed to face south in a shadeless area open to direct sunlight. As for the WI structures, the MS structures were spaced far enough apart to prevent any one structure from shading the next structure. The data from the MS structures provide a direct measure of a more severe (higher solar loading) location compared with Madison, Wisconsin (Winandy and Beaumont 1995). A typical exposure structure is shown in Figure 1.

3

Because of the slope of the Starkville test site, the east side of the MS structures was situated from 0.46 to 1.2 m off the ground, whereas the west side ranged from 0.15 to 0.46 m off the ground (Fig. 1). The WI structures were located on fairly flat ground and were approximately 0.5 m above grade.

All 10 exposure structures were identical. They were 3.7 m wide by 4.9 m long and constructed to simulate part of a typical multifamily attic–roof system in which U.S. Model Building Codes sometimes allow FR-treated plywood roof sheathing. To replicate this type of construction on a smaller scale, the 3.7-m-wide structures simulated in cross section the 1/8- to 3/8-span section of a 14.8-m span, 3:12 pitch roof system in both roof area and attic volume (Winandy and Beaumont 1995).

Each exposure structure was completely enclosed and unven-tilated. The four exterior walls were sheathed with 12-mm-thick, 200-mm-grooved Southern Pine siding attached to nominal 2- by 4-in. (standard 38- by 89-mm) wall studs. The exterior surfaces were coated with one coat of primer and two top coats of latex solid-color (light gray, almost white) stain or paint. The walls, floors, and roof system were not insulated.

The floor system was constructed from 9.5-mm-thick ply-wood floor sheathing and nominal 2- by 10-in. (standard 38- by 235-mm) joists. The rafters were made from nominal 2- by 6-in. (standard 38- by 140-mm) lumber, and the roof sheathing was 19-mm-thick plywood. Thick plywood roof sheathing was required because sixteen 100- by 550-mm openings were cut in each 1.2-m-wide by 2.4-m-long by 19-mm-thick panel to accommodate wood specimens used in the laboratory-to-field correlation studies. This study was intended to correlate thermal-induced degradation of FR-treated plywood in the field to matched FR-thermal-degradation rates derived in the laboratory (Barnes and others 1993).

The study variables were exposure location (Madison, Wis-consin, or Starkville, Mississippi), roof shingle color (black or white), and structure moisture content (heavily humidified or dry).

Recording of Temperatures To assess the effect of shingle color, the exposure structures were roofed with black or white CertainTeed XT–25 (Cer-tainTeed Corp., Blue Bell, PA) fiberglass shingles weighing 106 kg/square. The black and white shingles had reflectance values of 3.4% and 26.1%, respectively. Both black and white shingles had an emissivity rating of 0.91. These shin-gles were essentially identical to those used in Champaign, Illinois, to study the behavior of attics constructed and venti-lated in various ways (Rose 1992). Two WI structures had white shingles; the remaining three WI structures and all five MS structures had black shingles.

Three of the five WI exposure structures were each instru-mented with nine type-T thermocouples placed at various locations within the structure (Fig. 2). These structures were (1) a black-shingled structure that was not ventilated or humidified, (2) a black-shingled structure that was unventi-lated and artificially humidified from April through October to maintain >85% relative humidity for most of the diurnal cycle, and (3) a white-shingled structure that was not venti-lated or humidified. In Mississippi, all five structures were instrumented with thermocouples as shown in Figure 2. Two MS structures were humidified using a cold steam atomizing humidifier system such that the relative humidity was main-tained at >85% for most of the diurnal cycle. The interiors of the other three MS structures were kept dry.

The location and function of each thermocouple are de-scribed in Table 1. The thermocouples attached to the top ply of the roof sheathing (T3 and T4) were monitored because, theoretically, small roof systems such as the ones we tested have been reported to remain cooler than large roof struc-tures since air flow heats as it travels across the roof surface. Rose (1992) reported the effects of such a phenomenon on sheathing temperatures for a 9.2-m-wide roof system com-pared with a 12.8-m-wide system. The position of thermo-couple T7 allowed us to correlate rafter and sheathing tem-peratures. Thermocouple T8 acted as the external reference thermistor, with a rated accuracy of +0.2°C between −33°C and 48°C. Overall, the data from the thermocouples allowed us to correlate the roof temperatures of wood materials in various locations in a wood roof system with solar loads. Solar loads were monitored by the U.S. Weather Service at Mississippi State University in Starkville, Mississippi, and at the Dane County Regional Airport in Madison, Wisconsin.

To collect and record the temperature data at each location (Wisconsin and Mississippi) from each thermocouple loca-tion, two Campbell–Scientific (Logan, UT) model CR10 dataloggers and two model AM416, 32-channel multiplexers

Figure 1—Exposure structures at Mississippi Forest Products Laboratory, Mississippi State University, Starkville, Mississippi.

4

Table 1—Location and function of thermocouples in exposure structures a

Thermocouple Location Measurement

T0 Central, along back (north) wall Interior attic air temperature, 1 m above floor

T1 Central, along back (north) wall Interior attic air temperature, 2 m above floor

T2 Central, along back (north) wall Exterior air temperature, 2 m above grade

T3 Attached to top ply of plywood roof sheathing below roofing felt at 1/3 the rafter span, approximately 1.2 m from north (ridge) wall

Temperature of top ply of sheathing toward north wall

T4 Attached to top ply of plywood roof sheathing below roofing felt at 1/3 the rafter span, approximately 1.2 m from south (eave) wall

Temperature of top ply of sheathing toward south wall

T5 Attached to bottom ply of plywood roof sheathing at 1/3 the rafter span from the north (ridge) wall

Temperature of bottom ply of sheathing toward north wall

T6 Attached to bottom ply of plywood roof sheathing at 1/3 the rafter span from the south (eave) wall

Temperature of bottom ply of sheathing toward south wall

T7 Center of rafter at midspan Correlation of rafter and sheathing temperatures

T8 Datalogger External reference thermistor

aAdditional detail is shown in Figure 2.

~

3:12 pitch

Specimen

Roof felt

Plywoodsheathing

T7

T1

T0

Shingle

T3T5

T8Embeddedinsiderafter

~

North

38 by140 mmrafter

T6T4

Specimen

~

Datalogger

T8T5

T3

Shingles Roof feltPlywood

Rafters

T2

1.5 m

2.4 m

Test specimen

150 mm

Figure 2—Schematic cross section of instrumented field exposure structure slowing location of thermocouples (T0–T8) and channels for datalogger–multiplexer.

5

were used; one set (datalogger plus multiplexer) was used in Wisconsin and the other in Mississippi. The dataloggers had a reported accuracy of 0.2% over the service temperature range of −55°C to 85°C.

At each location, the datalogger and multiplexer were placed in a weather-sealed box located inside the middle structure of the five structures and located on a side wall approximately 1.83 m above the floor. Temperature data were collected every 5 min for each structure, and the datalogger was pro-grammed to calculate and record hourly average tempera-tures. Each week, the hourly data were downloaded and saved using a laptop computer. The set-up of the datalogger and laptop computer in Mississippi is depicted on the cover of this report. The Wisconsin set-up was nearly identical and was previously described in detail by Winandy and Beaumont (1995).

Measurement of Moisture Content At the Wisconsin site, moisture content was measured peri-odically at irregular intervals using an electrical resistance moisture content meter. The observations were collected during the day over the course of several years during all seasons. In Mississippi, moisture content was measured every 2 weeks over a 3-year period in four of the five exposure structures, except for one 4½-month period from late September 1998 to early February 1999.

Results and Discussion The fact that the field structures were neither ventilated nor insulated means that the thermal data hereafter reported are truly indicative only of such constructions. Larger structures might theoretically experience slightly higher temperatures, but to exactly what degree is unknown (Rose 1992). Finally, because the attic floors were not insulated and not heated or cooled, as would probably be the case in a traditional light-framed wood structure, the cooler floors would be expected to subtly affect radiation exchange between the attic and underlying areas. However, we believe that much practical information can be learned from studying these field struc-ture exposure temperature data.

We experienced thermocouple and multiplexer–datalogger problems in the WI heavily humidified (wet) black-shingled structure from January 1 to June 14 in 1996. We substituted the data from the WI dry black-shingled structure for this period because the dry and wet black-shingled structures in Wisconsin were found to have experienced very similar winter-to-spring temperatures in other years.

Sheathing Temperatures Top- and bottom-of-sheathing temperatures were monitored at the one-third and two-third points in the roof, midway between the eaves and the ridge. In both the MS and WI structures, the difference in recorded hourly temperatures

between the two top-of-sheathing thermocouples (one-third and two-third span) seldom exceeded 1°C. This same <1°C difference was also found for the two bottom-of-sheathing thermocouples. The small difference in temperature between these two matched locations across the span of the structure shows that little practical difference really existed in the structures. This might be related to the small size of the test structures or their lack of ventilation. Rose (1992) found that increased ventilation in pitched, cathedral ceiling systems increased the across-the-span temperature differential. How-ever, in a comparison of vented and unvented flat-roof sys-tems, Rose (1995) found few differences in sheathing tem-perature between the eaves and the ridge. Our small difference in temperatures across the length of the roof span seems to confirm that few real differences exist in sheathing temperature between the eaves and the ridge. Accordingly, the two top-of-sheathing and two bottom-of-sheathing tem-peratures were averaged, and the average hourly value is reported.

Exceedance Temperatures For the WI structures, the annual number of hours that the thermocouples recorded temperatures beyond various tem-perature limits were compiled into 5°C temperature bins. These 5°C bins (0°C to <5°C, 5°C to <10°C, …, 70°C to 75°C) are hereafter defined as “exceedance temperatures.” The value reported as the exceedance temperature for 70°C is thus the number of hours the temperature at that thermo-couple location equaled or exceeded 70°C, but is less than 75°C. Annual exceedance temperatures for the dry and wet black-shingled structures and dry white-shingled structure in Wisconsin, from 1992 to 1999, are given in Appendix A. Annual exceedance temperatures for the MS structures, from 1996 to 1999, are given in Appendix B. The exceedance temperature data were averaged over their 8- or 4-year his-tory to produce an annualized roof temperature history for each configuration.

Maximum Temperatures For the 4-year study at the Mississippi exposure site, maxi-mum “1-h average” temperatures recorded for black-shingled roofs in dry structures were 78°C and 63°C for the top and bottom plies of the plywood roof sheathing, respectively, and 58°C for the rafter. The maximum temperatures recorded for the matched WI structures were 75°C, 59°C, and 54°C, respectively. MS and WI black-shingled structures showed only small differences (3°C to 4°C) in maximum record temperatures.

The maximum temperatures recorded for the WI white-shingled structure over the 8-year period were 64°C, 53°C, and 49°C for the top and bottom plies of the roof sheathing and the rafter, respectively. This clearly shows that the use of white shingles, as opposed to black shingles, can

6

dramatically lower the maximum temperatures—by 5°C to 11°C—in wood materials used in roof systems.

The maximum temperatures recorded in the 4-year Missis-sippi study for black-shingled roofs in heavily humidified structures were cooler than those for matched dry structures. For the wet black-shingled MS structures, maximum tem-peratures for the top and bottom plies of the roof sheathing and the rafter were 74°C, 58°C, and 54°C, respectively. The maximum temperatures of black-shingled roofs in heavily humidified structures in the 4-year Wisconsin study were cooler than those for matched dry WI structures. For the wet black-shingled WI structures, maximum temperatures for the top and bottom plies of the roof sheathing and the rafter were 74°C, 58°C, and 52°C, respectively. The maximum tempera-tures of wet MS and dry and wet WI structures were virtually identical. Daily 1-h maximum temperatures and annualized roof temperature data for Wisconsin exhibited similar values and trends to previously reported data from a 3-year study in Wisconsin (Winandy and Beaumont 1995).

Annual Temperature Trends The average 8- or 4-year temperature histories for each thermocouple in each exposure structure are plotted for the WI structures in Figures 3 to 5 and for the MS structures in Figures 6 and 7. Note the general similarity in relative form of the exceedance temperature distributions between each exposure site, roof color, and attic humidity configuration. As expected, the MS structures experienced many more hours of exposure to high air temperatures than did the WI structures. However, although air temperatures were often warmer in Mississippi than in Wisconsin, the differences in annual maximum temperatures of the top and bottom of the plywood roof sheathing were very similar. We observed that the higher portions of the roof sheathing temperature histo-ries were clearly controlled by solar radiation and not the outside air temperature (Figs. 3 to 7), which reaffirms the earlier observations of Winandy and Beaumont (1995) and the predictions of the TenWolde (1997) model. Figures 3 to 7 also show how internal rafter temperature is nearly coinci-dent with inside attic air temperature and how both track relatively closely with outside air temperature. Thus, al-though solar radiation may control sheathing temperature, attic air space or rafter temperature is strongly influenced by outside air temperatures during most of the year.

Figures 8 to 11 show a direct comparison of temperatures recorded by individual thermocouples for each building configuration. Note that for both Mississippi and Wisconsin, the wet (heavily humidified) roof systems were generally similar in temperature to dry roof systems except during the hottest days or the highest portions of the temperature range. In the high temperature range, the thermocouples in nearly every location within the wet MS structures were usually 3°C to 4°C cooler than matched thermocouples in the dry MS structures. We believe that this was caused by evaporative

cooling of the sheathing and rafter lumber during those warmest periods of each hot day. Although both the MS and WI wet structures experienced this same characteristic “evaporative cooling,” the matched wet and dry MS struc-tures seemed to experience an even more pronounced devia-tion in thermocouple temperatures. We expect that this oc-curred because Mississippi has a warmer, more humid climate, and the MS exposure structures were subject to longer periods in the warmer, more humid temperature range,

1

10

100

1000

10000

Ave

rage

hou

rs/y

ear

0 10 20 30 40 50 60 70 80 Exceedence temperature (oC)

RafterBottom plyTop plyOutside airInside air

WI, black, dry

Figure 3—Annualized 8-year temperature history for dry black-shingled structure in Wisconsin.

1

10

100

1000

10000

Ave

rage

hou

rs/y

ear



WI, black, wet

Figure 4—Annualized 4-year temperature history for wet black-shingled structure in Wisconsin.

1

10

100

1000

10000

Ave

rage

hou

rs/y

ear



WI, white, dry

Figure 5—Annualized 8-year temperature history for dry white-shingled structure in Wisconsin.

7

where measurable “evaporative cooling” would be more likely.

The warmer climate of Mississippi is readily apparent in the plot of the annualized 8- or 4-year average temperatures (Fig. 12a). Note that the temperature measured by each thermocouple was generally higher (by 10°C to 14°C) in the MS structures compared with the WI structures. Furthermore, temperatures of wet and dry structures were generally simi-lar, except for the previously discussed deviation at the high-est temperatures (Fig. 12b). In both the MS and WI

structures, the maximum plywood sheathing, rafter, and internal attic air temperatures during the monitoring periods were more similar than dissimilar, as earlier claims had suggested (ASTM 1988).

The minimum temperatures reveal the real difference in climatic influences for Mississippi and Wisconsin (Fig. 12c). Note that the annual minimum temperatures of wood compo-nents in WI structures (–35° to –40°C) were as much as 20°C colder than the minimum temperatures of MS structures (–20°C). Also note that the minimum temperatures of WI


MS, black, dry

1000

10000A

vera

ge h

ours

/yea

r

1

10

100


Figure 6—Annualized 4-year temperature history for dry black-shingled structure in Mississippi.


1

10

100

1000

10000

Ave

rage

hou

rs/y

ear


MS, black, wet

Figure 7—Annualized 4-year temperature history for wet black-shingled structure in Mississippi.

1

10

100

1000

10000

Ave

rage

hou

rs/y

ear


WI, white, dry

WI, black, wet

WI, black, dry

MS, black, wet

MS, black, dry

Attic

Figure 8—Attic air temperature history for test structures.

1

10

100

1000

10000

Ave

rage

hou

rs/y

ear


WI, white, dry

WI, black, wet

WI, black, dry

MS, black, wet

MS, black, dry

Roof sheathing, top

Figure 9—Air temperature history for top surface of plywood roof sheathing.

1

10

100

1000

10000

Ave

rage

hou

rs/y

ear


WI, white, dry

WI, black, wet

WI, black, dry

MS, black, wet

MS, black, dry

Roof sheathing, bottom

Figure 10—Air temperature history for bottom surface of plywood roof sheathing.

1

10

100

1000

10000

Ave

rage

hou

rs/y

ear


WI, white, dry

WI, black, wet

WI, black, dry

MS, black, wet

MS, black, dry

Internal roof rafter

Figure 11—Air temperature history for internal rafters.

8

Figure 12—Average, maximum, and minimum 8- or 4-year temperatures for exposure structures.

(a) 8- or 4-year averages

0

5

10

15

20

25

30

0

-45 Inside Outside Sheathing

top Sheathing

bottom Rafter

(c) 8- or 4-year minimums

-40

-35

-30

-25

-20

-15

-10

-5

10

5

0

(b) 8- or 4-year maximums

10

20

30

40

50

60

70

80

90

WI, white, dry

WI, black, dry WI, black, wet

MS, black, dry MS, black, wet

Tem

pera

ture

(° C

)

9

black- and white-shingled structures (Fig. 12c) were very similar to each other, whereas the maximum temperatures of these structures were very different from each other (Fig. 12b).

Annual outside air temperatures in Madison, Wisconsin, were distinctly bimodal. Thermocouple T2 readings showed two distinct peaks at 10°C to 15°C and –5°C to 0°C, with noticeably fewer values in the 5°C to 10°C range (Tables 2 to 4). The other thermocouple readings (attic air, top and bottom of plywood roof sheathing, and rafter) also showed signs of bimodality, but not as obviously. These findings are evident in the data for the WI structures by individual year and location–interior humidity configuration (Appendix A). The temperature histories of all thermocouple locations in the MS structures were unimodal (Appendix B). The Mississippi data did not exhibit the bimodal nature of the Wisconsin data. Tables 5 and 6 show the unimodal nature of the thermal loads for the MS structures.

Peak Temperature Trends The temperatures achieved at each thermocouple location during the hottest 7-consecutive-day period over the 8- or 4-year history of each exposure structure are shown in Figures 13 and 14; the influence of exposure site and attic humidity during this time is shown in Figure 15. The individ-ual “peak” temperatures recorded for both the dry black-shingled and dry white-shingled WI structures during the 8-year exposure period were similar to those recorded previ-ously for a 3-year exposure study (Winandy and Beaumont 1995). Although the wood components under the white roofing material were much cooler on “peak” days than those under the black roofing material, the differences between sheathing temperatures of the dry black-shingled and dry white-shingled WI attics were minimal.

Table 2—Eight-year average times at given temperatures for WI dry, black-shingled structure

Time (h) at given temperature at various locations

Structure Roof sheathing Temper–

ature (°C) Inside Outside Top Bottom Rafter

–40 0 –– 1 1 1

–35 7 6 9 8 7

–30 13 15 18 14 13

–25 48 51 71 55 51

–20 149 162 172 156 149

–15 256 314 289 267 259

–10 548 622 595 571 561

–5 1,063 1,220 1,057 1,064 1,062

0 1,276 1,340 1,216 1,246 1,269

5 987 990 948 967 977

10 1,034 1,131 1,003 1,030 1,035

15 1,063 1,297 964 1,038 1,050

20 846 1,040 652 771 834

25 509 486 385 480 514

30 399 88 308 365 402

35 322 4 258 322 322

40 186 –– 225 240 194

45 55 –– 210 134 62

50 5 –– 168 35 6

55 –– –– 121 4 ––

60 –– –– 70 –– ––

65 –– –– 23 –– ––

70 –– –– 5 –– ––

75 –– –– 0 –– ––

Table 3—Four-year average times at given temperatures for WI wet, black-shingled structure




–40 1 –– 2 1 1

–35 7 6 9 8 8

–30 12 14 14 12 11

–25 43 49 61 48 43

–20 138 161 160 151 142

–15 240 282 272 246 239

–10 523 554 568 555 532

–5 1,073 1,261 1,081 1,071 1,070

0 1,246 1,352 1,175 1,208 1,246

5 961 958 935 942 943

10 1,056 1,164 1,051 1,038 1,033

15 1,090 1,282 978 1,038 1,061

20 911 1,070 668 824 916

25 565 501 400 501 560

30 424 105 306 383 428

35 325 7 262 321 340

40 134 –– 230 255 170

45 17 –– 209 136 27

50 1 –– 172 26 2

55 –– –– 128 3 ––

60 –– –– 65 –– ––

65 –– –– 20 –– ––

70 –– –– 4 –– ––

75 –– –– –– –– ––

10

As discussed previously in this report, wood components in the matched WI and MS structures experienced similar an-nual maximum temperatures, but MS structures were ex-posed to those peak temperature ranges for more hours each day. This result is clearly evident in the comparison of tem-peratures recorded over the hottest 7-consecutive-day period over the 8- and 4-year histories of the respective Wisconsin and Mississippi exposure sites (Figs. 13 and 14, respec-tively). The same scenario is evident in the temperature histories of wood components in the heavily humidified MS and WI structures. Like the temperatures of the dry struc-tures, the individual annual peak temperatures of the wet black-shingled structures in Wisconsin (Fig. 13) and Missis-sippi (Fig. 14) were similar, but the daily peak temperatures in the MS structures were higher. These results suggest that the benefits and limitations of attic ventilation during warm summer periods are similar regardless of whether the build-ings are located in the North or the South.

Many important trends can be summarized by analyzing the wood component and attic air temperatures over the hottest 7-consecutive-day period over the Wisconsin and Mississippi histories. Attic air temperatures were clearly higher in the MS structures than in the WI structures (Fig. 15). The effect of evaporative cooling is illustrated by the fact that attic air temperature in the MS wet black-shingled structure was similar to that in the WI dry black-shingled structure, and attic air temperature in the WI wet black-shingled structure was similar to that in the WI dry white-shingled structure. On the contrary, the top-of-sheathing temperatures of the dry and wet structures in both Mississippi and Wisconsin were very similar (Fig. 15). This seems to reinforce the idea that sheath-ing temperatures, especially temperatures at the top of the sheathing, are primarily controlled by solar radiation.

The temperatures of the bottom of the plywood sheathing fell between the temperatures recorded for the attic air and the top of the sheathing (Fig. 15). We observed that bottom-of- sheathing temperatures followed a thermal history that

Table 4—Eight-year average times at given temperatures for WI dry, white-shingled structure




–40 0 –– 2 1 1 –35 7 6 9 8 7 –30 12 15 18 14 13 –25 47 50 74 55 50 –20 143 161 177 159 150 –15 258 308 296 273 262 –10 559 614 614 588 574

–5 1,074 1,213 1,111 1,097 1,085 0 1,308 1,333 1,248 1,288 1,307 5 993 996 976 985 987

10 1,058 1,145 1,045 1,062 1,057 15 1,102 1,278 1,003 1,061 1,086 20 888 1,061 700 821 868 25 546 492 408 490 538 30 428 91 318 397 427 35 266 3 271 287 266 40 73 –– 227 148 82 45 6 –– 161 31 8 50 –– –– 83 2 –– 55 –– –– 23 –– –– 60 –– –– 5 –– –– 65 –– –– –– –– –– 70 –– –– –– –– –– 75 –– –– –– –– ––

Table 5—Four-year average times at given temperatures for MS dry, black-shingled structure




–40 –– –– –– –– –– –35 –– –– –– –– –– –30 –– –– –– –– –– –25 –– –– –– –– –– –20 1 1 8 4 2 –15 13 14 29 15 13 –10 55 65 107 72 57

–5 202 243 331 258 215 0 552 591 609 562 544 5 882 1,080 857 866 870

10 1,078 1,273 1,043 1,048 1,057 15 1,150 1,423 1,159 1,108 1,139 20 1,596 1,925 1,398 1,532 1,563 25 1,070 1,193 641 866 1,036 30 639 783 421 581 647 35 494 176 355 482 493 40 507 –– 348 399 466 45 398 –– 338 418 445 50 130 –– 284 356 211 55 2 –– 310 190 11 60 –– –– 272 13 –– 65 –– –– 194 –– –– 70 –– –– 64 –– –– 75 –– –– 2 –– ––

11

seemed to be partially controlled by both the attic air and top-of-sheathing temperatures. Rafter temperatures also appeared to mirror the thermal history of bottom-of-sheathing temperatures. Rafter temperatures in the wet black-shingled MS structure fell between those of the MS dry black-shingled and WI dry black-shingled structures (Fig. 15). These observations reinforce the idea that the bottom-of-sheathing and internal rafter temperatures are jointly influenced by outside air and attic air temperatures and by solar radiation.

As discussed previously in this report, engineering design standards for wood (AF&PA 1997) require a strength prop-erty adjustment for sustained exposures above 37.8°C and greater adjustment for prolonged exposures up to 65.6°C. Average annualized exposures at various critical tempera-tures were derived using a nonparametric analysis. These values (Table 7) show the relative time the various wood building components spent in three AF&PA temperature regimes: (1) ≥37.8°C but <51.7°C, (2) ≥51.7°C but <65.6°C, and (3) ≥65.6°C. The data show that roof plywood sheathing and roof rafters are in these important temperature ranges for a significant portion of time. Recently derived models for the

residual serviceability of untreated wood roofing material have shown that such thermal loading might account for a 4% loss in strength for each 10 years of exposure in Madi-son, Wisconsin (Winandy 1998).

Moisture Content Rose (1992) reported that when a good vapor barrier was maintained between warm living spaces and cooler attics, the moisture content (MC) of wood components used in flat- or cathedral-ceiling attics ranged from 8% to 22%, averaged between 13% and 15%, and was higher than 18% for less than 20 to 50 hours per year. These findings were similar regardless of whether the roof systems were ventilated or unventilated. However, Rose found that when the ceiling vapor barrier was broken, lumber in unventilated cathedral roof cavities could have more than 30% moisture content for more than 200 h/year. In our study, none of the MC meas-urements taken at the Wisconsin site was higher than 20% or lower than 6%. In winter, MC tended to range from 13% to 16% and in summer, from 8% to 12%. At the Mississippi site, wood MC was lowest during summer and highest during winter. In dry roof systems, sheathing MC varied from 1.5% in summer to 7.5% in winter; in artificially humidified roof systems, sheathing MC varied from 4% in summer to 17% in winter (Fig. 16). Thus, the results from the WI structures, which were not humidified during cold weather periods, and the MS structures, which were humidified all year long, affirm the trends in wood MC previously reported by Rose (1992).

Concluding Remarks The results of this research study are summarized as follows:

• The annual 1-h maximum temperatures of various wood components were similar in Mississippi and Wisconsin roof systems; these temperatures were only 3°C to 4°C higher in the MS structures.

• Although the annual maximum and the form of the re-corded exceedance temperatures were similar in the MS and WI exposure structures, the MS structures experienced temperatures in the higher range for many more hours per year compared to matched WI structures. Temperatures of wood components in the MS structures were generally 5°C to 10°C warmer than those of matched WI structures.

• Black-shingled roof systems tended to be 5°C to 10°C warmer on sunny afternoons compared with white-shingled systems.

• Moisture content of plywood roof sheathing in dry MS structures varied from 1.5% to 7.5% between summer and winter; moisture content of sheathing in wet (heavily hu-midified) MS structures varied from 4% to 17%. Moisture content of sheathing in dry WI structures varied from 6% to 13%.

Table 6—Four-year average times at given temperatures for MS wet, black-shingled structure




–40 –– –– –– –– –– –35 –– –– –– –– –– –30 –– –– –– –– –– –25 –– –– –– –– –– –20 –– 1 5 1 –– –15 8 14 14 11 9 –10 23 65 97 36 26

–5 229 243 316 254 232 0 578 591 601 568 559 5 915 1,080 871 867 887

10 1,130 1,273 1,032 1,058 1,079 15 1,213 1,423 1,160 1,130 1,169 20 1,686 1,925 1,425 1,581 1,632 25 1,028 1,193 686 961 1,074 30 625 783 456 646 652 35 557 176 378 473 524 40 480 –– 366 471 508 45 283 –– 324 430 339 50 14 –– 312 256 78 55 –– –– 312 22 –– 60 –– –– 250 –– –– 65 –– –– 145 –– –– 70 –– –– 18 –– –– 75 –– –– –– –– ––

12

20

30

40

50

60

70

80

July 1999

(a) WI, black, dry

24 25 282726 30 3129

Tem

pera

ture

(o C

)

20

30

40

50

60

70

80(b) WI, black, wet

20

30

40

50

60

(c) WI, white, dry

70

80


Figure 13—Temperatures at each thermocouple location during 7 hottest consecutive days in 8-year exposure of WI structures.

13

• Temperatures at the top of the roof sheathing were con-trolled by solar gain, not outside air or attic air tempera-tures.

• Temperatures at the bottom of the roof sheathing were usually controlled by solar gain, except on a few of the hottest days, when sheathing temperatures were also influ-enced by outside air or attic air temperatures.

• Rafter temperatures were usually controlled by attic air temperatures, except on a few of the hottest days, when they were also influenced by solar radiation.

• On hot days in heavily humidified structures, evaporative cooling is responsible for lower temperatures in plywood sheathing and wood rafters.

• The major difference in the temperature of wood compo-nents used in attics in the northern exposure (Wisconsin) compared with those used in the southern exposure (Mis-sissippi) was in minimum temperatures, which were as much as 20°C lower in the WI structures.

• Annual outside air temperature is bimodal in Wisconsin, with two distinct peaks at 10°C to 15°C and −5°C to 0°C. Annual outside air temperature is unimodal in Mississippi, averaging between 10°C and 15°C.

20

30

40

50

60

70

80Te

mpe

ratu

re (

o C)

20

30

40

50

60

70

80

June 1998

(a) MS, black, dry

20 21 242322 26 2725

Rafter (b) MS, black, wetBottom plyTop ply Outside airInside air

Figure 14—Temperatures at each thermocouple location during 7 hottest consecutive days in 4-year exposure of MS structures.

14

RafterBottom plyTop plyOutside airInside air (a) Attic air

20

30

40

50

60

70

80

20

30

40

50

60

70

80

1 2 3 4 5 6 7 8Day

(d) Rafter

MS, black, wetMS, black, dryWI, white, dryWI, black, wetWI, black, dry

20

30

40

50

60

70

80

20

30

40

50

60

70

80 (c) Sheathing bottom

Tem

pera

ture

(o C

)

(b) Sheathing top

Figure 15—Influence of exposure site and attic humidity during 7 hottest consecutive days in exposure period.

15

Ave

rage

moi

stur

e co

nten

t (%

)

2

6

10

14

18

22

Mar15

Mar30

Apr10

Apr25

May23

Jun2

6

Jul25

Aug31

Oct06Nov

22

Jan0

9Feb

27

Apr25

Jun1

1

Aug09

Sep23

Nov04

Dec05

Jan2

3

Feb27

Apr15

Jul09

Aug19

Dec02

Feb18

Mar27

Apr29

Sep29

Feb19

Apr23

Jun2

5

Aug02

Sep09

MS, black, wet

MS, black, dry

Ave

rage

moi

stur

e co

nten

t (%

)

Mar

15

Mar

31Apr

18

May

18

Jun2

6

Aug09

Sep21

Nov08

Jan0

9

May

19Ju

l22

Sep23

Nov11

Dec17

Feb1

3

Apr15

Nov11

Feb1

0

Mar

27

Jun2

4

Feb1

9

May

06

Jul12

Aug30

Mar

141

2

3

4

5

6

7

8

Figure 16—Moisture content history for dry and wet MS black-shingled structures.

16

Acknowledgments The authors would like to acknowledge the financial assis-tance of the USDA Competitive Grants program (grant no. 93–02444) and the New Jersey Department of Community Affairs. We also acknowledge the technical assistance of Louis Watson at Mississippi State University and Mike Grambsch and Earl Geske at Forest Products Laboratory, who monitored and programmed the temperature datalogging equipment. Rita Simonsen of the Forest Products Laboratory archived 8 years worth of field data. Finally, we wish to thank the CertainTeed Corporation for donation of the roof-ing materials used for the outdoor simulated-attic structures.

References

AF&PA. 1997. National design specification for wood construction. An American National Standards Insti-tute/American Forest and Paper Association Standard. ANSI/AF&PA NDS–1997. Washington, DC: American Forest & Paper Association.174 p.

APA. 1989. Fire-retardant-treated plywood: Prediction of performance. Tacoma, WA: American Plywood Association.

ASTM. 1988. Effect of fire retardant treatments (FRT) on the strength properties of wood. In: Minutes of workshop;

1988 April 26; Madison, WI: ASTM D7.06.04 Section: Fire performance of wood. U.S. Department of Agriculture, For-est Service, Forest Products Laboratory. 7 p.

ASTM. 1999. Standard practice for calculating bending strength design adjustment factors for fire-retardant-treated plywood roof sheathing. ASTM Annual Book of Standards, ASTM D6305–98. West Conshohocken, PA: American Society for Testing and Materials.

Barnes, H.M.; Winandy, J.E.; Conners, T.E. 1993. Labo-ratory/field correlations for FRT plywood. USDA National Research Initiative, competitive grant proposal for natural resources and the environment, Grant no. 93–02444. Wash-ington, DC: U.S. Department of Agriculture, National Re-search Initiative.

Blackenstowe, D.E. 1987. Comparison of white versus black surfaces for energy conservation. In: Proceedings, 8th conference on roofing technology; 1897 April 16–17. Gaithersburg, MD: National Bureau of Standards.

Heyer, O.C. 1963. Study of temperature in wood parts of houses throughout the United States. Res. Note FPL–RN–012. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory.

Table 7—Duration of exposure of each test structure in high-temperature exposure ranges in current building design a

Site Shingle Humidity

Sheathing or rafter location

≥37.8°C but

<51.7°C

≥51.7.°C but

<65.6°C ≥65.6°C

WI Black Dry Top 608 302 23

Bottom 526 20 —

Rafter 389 2 —

Wet Top 609 308 20

Bottom 537 13 —

Rafter 335 1 —

White Dry Top 546 73 —

Bottom 292 1 —

Rafter 190 — —

MS Black Dry Top 932 790 234

Bottom 1,160 428 —

Rafter 1,232 122 —

Wet Top 968 783 144

Bottom 1,195 176 —

Rafter 1,148 15 — aAF&PA 1997.

17

Lebow, P.K.; Winandy, J.E. 1999. Verification of kinetics-based model for long-term effects of fire-retardants on bend-ing strength at elevated temperatures. Wood and Fiber Science. 31(1): 49–61.

LeVan, S.L.; Collet, M. 1989. Choosing and applying fire-retardant-treated plywood and lumber for roof designs. Gen. Tech. Rep. FPL–GTR–62. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory. 11 p.

NAHB. 1990. Home builders guide to fire retardant treated plywood: Evaluation, testing, and replacement. Washington, DC: National Research Center of the National Association of Home Builders. 65 p.

NBS. 1979. Summer cooling and whole house ventilation. NBS special pub. 548. Reppart, M.H., ed. Gaithersburg, MD: National Bureau of Standards, U.S. Department of Com-merce.

Ozkan, E. 1993. Surface and inner temperature attainment of flat roof systems in hot–dry climate. Riyadh, Saudi Arabia: King Saud University, Riyadh.

Rose, W.B. 1992. Measured values of temperature and sheathing moisture content in residential attic assemblies. In: Geshwiler, M., ed. Thermal performance of the exterior envelopes of building. In: Proceedings, ASHRAE/DOE/ BTECC conference; 1992 December 1–10; Clearwater Beach, FL. Atlanta, GA: American Society of Heating, Re-frigerating, and Air-Conditioning Engineers: 379–390.

Rose, W.B. 1995. Temperature and sheathing moisture content in residential attics. Unpub. rep., Champaign, IL: University of Illinois, Building Research Center.

TenWolde, A. 1988. The FPL roof temperature model. In: Effects of fire retardant treatments on strength properties of wood; Executive summary, ASTM D7 workshop; 1988 April. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Laboratory.

TenWolde, A. 1997. FPL roof temperature and moisture model. Res. Pap. FPL–RP–561. Madison, WI: U.S. Depart-ment of Agriculture, Forest Service, Forest Products Labora-tory.

Wilkes, K.E. 1989. Model for roof thermal performance. Oak Ridge National Laboratory Rep. ORNL/CON–274. Oak Ridge, TN: Office of Scientific and Technical Information.

Winandy, J.E. 1998. Using kinetics-based models to address serviceability concerns for fire-retardant-treated wood at elevated in-service temperatures. In: Proceedings, 5th World conference on timber engineering; 1998 August 16–21: Montreux, Switzerland. Lausanne, Switzerland: Swiss Fed-eral Institute of Technology.

Winandy, J.E. 2000. Serviceability modeling—Predicting and extending useful service life of FRT plywood roof sheathing. In: Proceedings, International Research Group for Wood Preservation, IRG/WP 00–20210, Stockholm, Swe-den.

Winandy, J.E.; Beaumont, R. 1995. Roof temperatures in simulated attics. Res. Pap. FPL–RP–543. Madison, WI: U.S. Department of Agriculture, Forest Service, Forest Products Laboratory.

Appendix A—Exceedance Temperature Data for WI Structures The data in Tables 8 to 27 show exceedance temperature data for field exposure structures in Madison, Wisconsin, from 1992 to 1999.

Table 8—Exceedance temperatures in WI dry, black-shingled structure, 1992




−40 — — — — — –35 — — — — — –30 — — 12 3 — –25 32 25 32 33 35 –20 56 67 60 58 52 –15 172 183 244 196 176 –10 496 539 558 526 516

–5 1,230 1,453 1,238 1,250 1,243 0 1,589 1,714 1,485 1,542 1,565 5 1,035 961 1,023 1,023 1,036

10 1,020 1,149 977 1,020 1,030 15 1,075 1,427 935 1,017 1,032 20 772 929 559 677 752 25 467 305 369 445 487 30 374 32 292 349 379 35 310 — 258 297 314 40 135 — 204 225 142 45 21 — 204 104 25 50 — — 158 19 — 55 — — 103 — — 60 — — 61 — — 65 — — 12 — — 70 — — — — — 75 — — — — —

18





–40 — — — — — –35 — — — — — –30 — — 3 — — –25 34 17 66 50 38 –20 143 152 183 154 144 –15 278 385 277 280 277 –10 592 703 642 631 603

–5 1,014 1,165 1,030 1,022 1,026 0 1,542 1,585 1,442 1,487 1,518 5 881 875 816 848 872

10 983 1,136 988 999 979 15 1,104 1,283 975 1,061 1,084 20 865 1,046 676 776 882 25 532 381 396 496 530 30 405 32 325 358 396 35 276 — 265 330 289 40 84 — 223 183 92 45 25 — 209 63 28 50 2 — 131 19 2 55 — — 68 3 — 60 — — 27 — — 65 — — 13 — — 70 — — 5 — — 75 — — — — —





–40 1 — 5 4 2 –35 27 22 35 29 29 –30 54 65 68 58 53 –25 98 124 110 110 104 –20 173 211 172 172 174 –15 268 300 305 276 263 –10 561 619 633 592 581

–5 860 914 842 860 858 0 986 1,088 999 998 984 5 1,075 1,189 998 1,035 1,065

10 1,045 1,134 1,012 1,044 1,052 15 1,223 1,528 1,059 1,164 1,198 20 832 1,007 663 765 832 25 554 486 414 505 555 30 444 69 314 388 432 35 335 4 264 334 332 40 186 — 246 260 204 45 35 — 225 132 39 50 3 — 183 31 3 55 — — 129 3 — 60 — — 59 — — 65 — — 22 — — 70 — — 2 — — 75 — — 1 — —





–40 — — — — — –35 — — — — — –30 2 — 8 5 3 –25 55 45 95 66 56 –20 204 188 250 217 204 –15 330 402 331 345 337 –10 591 828 625 604 593

–5 1,078 1,174 1,032 1,057 1062 0 1,187 1,172 1,127 1,161 1,202 5 1,032 985 980 1,016 1,019

10 902 1,000 860 897 906 15 825 1,024 839 836 819 20 923 1,109 761 827 914 25 550 635 387 537 555 30 413 187 325 343 410 35 305 11 235 315 295 40 242 — 219 244 250 45 109 — 213 199 120 50 12 — 170 78 15 55 — — 134 13 — 60 — — 115 — — 65 — — 43 — — 70 — — 11 — — 75 — — — — —





–40 2 — 6 3 3 –35 24 22 28 26 24 –30 32 44 32 31 29 –25 82 86 125 92 84 –20 235 287 256 245 236 –15 316 435 337 318 327 –10 672 712 721 702 684

–5 1,154 1,337 1,094 1,131 1,147 0 1,123 1,187 1,041 1,078 1,112 5 828 872 884 850 815

10 1,149 1,120 1,171 1,134 1,135 15 1,081 1,206 867 1,030 1,097 20 756 953 586 703 742 25 430 459 366 421 436 30 354 54 269 335 363 35 326 10 248 318 331 40 172 — 222 220 165 45 48 — 203 131 — 50 — — 157 16 — 55 — — 91 — — 60 — — 67 — — 65 — — 13 — — 70 — — — — — 75 — — — — —

19





–40 — — — — — –35 — — — — — –30 6 — 9 6 5 –25 40 53 65 46 46 –20 146 162 181 152 145 –15 229 270 255 241 235 –10 521 623 611 544 543

–5 1,213 1,385 1,185 1,216 1,218 0 1,369 1,441 1,287 1,333 1,347 5 934 837 915 909 927

10 1,025 1,146 945 1,019 1,035 15 996 1,339 989 1,005 986 20 869 1,023 605 773 842 25 496 418 372 483 498 30 384 63 317 365 388 35 304 — 252 306 295 40 181 — 225 213 197 45 41 — 192 124 47 50 6 — 158 21 6 55 — — 117 4 — 60 — — 57 — — 65 — — 18 — — 70 — — 5 — — 75 — — — — —





–40 — — — — — –35 — — — — — –30 — — 3 — — –25 15 16 27 17 15 –20 81 75 116 91 84 –15 148 206 157 150 151 –10 380 349 427 384 380

–5 949 1,157 989 960 950 0 1,278 1,339 1,230 1,254 1,274 5 1,054 1,079 961 1,023 1,045

10 1,169 1,286 1,185 1,164 1,159 15 1,117 1,342 1,063 1,133 1,126 20 918 1,167 690 848 882 25 511 656 379 471 520 30 420 88 296 393 438 35 378 — 278 355 364 40 272 — 243 298 287 45 69 — 224 181 78 50 1 — 204 38 7 55 — — 169 — — 60 — — 85 — — 65 — — 27 — — 70 — — 7 — — 75 — — — — —





–40 — — — — — –35 5 2 8 6 5 –30 9 11 11 8 10 –25 31 44 49 28 28 –20 150 152 160 159 156 –15 310 330 404 329 309 –10 568 604 546 581 584

–5 1,007 1,176 1,042 1,016 993 0 1,134 1,195 1,113 1,117 1,147 5 1,054 1,119 1,009 1,029 1,033

10 980 1,079 888 960 986 15 1,080 1,228 981 1,056 1059 20 832 1,085 675 801 824 25 532 550 394 479 534 30 401 178 326 385 408 35 344 7 261 322 352 40 219 — 221 278 211 45 91 — 208 139 106 50 13 — 179 59 15 55 — — 156 8 — 60 — — 85 — — 65 — — 34 — — 70 — — 10 — — 75 — — — — —

Table 16—Exceedance temperatures in WI wet, black-shingled structure, 1996




–40 2 — 6 3 3 –35 24 22 28 26 24 –30 32 44 32 31 29 –25 82 88 124 94 85 –20 233 285 255 249 234 –15 324 433 339 317 325 –10 672 708 714 712 689

–5 1,135 1,342 1,095 1,121 1,133 0 1,129 1,191 1,037 1,067 1,113 5 817 854 849 826 791

10 1,086 1,098 1,156 1,057 1,050 15 1,128 1,199 889 1,062 1,136 20 785 913 612 740 810 25 473 468 380 449 470 30 377 123 276 340 381 35 324 16 248 311 318 40 151 — 219 232 176 45 10 — 210 135 17 50 — — 155 12 — 55 — — 99 — — 60 — — 51 — — 65 — — 10 — — 70 — — — — — 75 — — — — —

20





–40 — — — — — –35 — — — — — –30 5 — 9 6 4 –25 44 52 57 47 44 –20 138 163 173 160 142 –15 240 266 266 246 237 –10 529 613 589 558 522

–5 1,188 1,406 1,211 1,224 1,235 0 1,377 1,428 1,265 1,301 1,348 5 915 823 886 870 879

10 977 1,143 955 983 952 15 1,030 1,364 975 966 971 20 940 1,033 642 825 977 25 598 414 411 532 582 30 425 55 302 386 437 35 258 — 265 319 288 40 87 — 216 209 128 45 9 — 199 110 14 50 — — 163 16 — 55 — — 104 2 — 60 — — 56 — — 65 — — 12 — — 70 — — 4 — — 75 — — — — —





–40 — — — — — –35 — — — — — –30 — — 3 1 –25 16 15 25 21 15 –20 82 74 113 93 89 –15 146 201 155 154 149 –10 369 353 431 386 373

–5 992 1,141 1,010 984 961 0 1,224 1,350 1,197 1,244 1,267 5 1,020 1,066 958 994 1,014

10 1,147 1,289 1,165 1,125 1,116 15 1,136 1,345 1,072 1,119 1,107 20 1,005 1,242 723 886 980 25 592 614 394 493 602 30 464 70 307 430 463 35 400 — 268 335 404 40 154 — 266 311 203 45 13 — 216 162 17 50 — — 187 22 — 55 — — 167 — — 60 — — 78 — — 65 — — 22 — — 70 — — 3 — — 75 — — — — —





–40 — — — — — –35 5 2 8 7 6 –30 9 11 11 8 9 –25 29 40 38 31 27 –20 100 123 99 101 103 –15 248 229 326 268 243 –10 522 542 536 565 543

–5 975 1,154 1,009 955 949 0 1,255 1,440 1,199 1,221 1,255 5 1,091 1,090 1,047 1,077 1,088

10 1,015 1,127 929 988 1,012 15 1,067 1,220 976 1,005 1,028 20 915 1,093 695 845 895 25 598 507 416 529 586 30 430 172 337 374 431 35 318 10 265 320 349 40 145 — 218 269 171 45 34 — 212 135 58 50 4 — 182 53 7 55 — — 140 9 — 60 — — 73 — — 65 — — 36 — — 70 — — 8 — — 75 — — — — —

Table 20—Exceedance temperatures in WI dry, white-shingled structure, 1992




–40 — — — — — –35 — — — — — –30 — — 12 2 — –25 31 25 36 34 35 –20 54 66 64 59 53 –15 168 184 249 200 175 –10 504 538 574 538 524

–5 1,271 1,437 1,284 1,281 1,283 0 1,622 1,713 1,532 1,585 1,603 5 1,035 976 1,059 1,047 1038

10 1,037 1,149 1,019 1,057 1,059 15 1,124 1,407 976 1,058 1,110 20 793 937 615 742 772 25 505 316 382 444 477 30 409 36 303 379 412 35 199 — 267 256 206 40 32 — 210 89 37 45 — — 135 13 — 50 — — 59 — — 55 — — 8 — — 60 — — — — — 65 — — — — — 70 — — — — — 75 — — — — —

21





–40 — — — — — –35 — — — — — –30 — — 4 — — –25 33 19 69 47 39 –20 142 147 187 160 146 –15 278 378 288 281 280 –10 616 703 662 659 627

–5 1,054 1,179 1,114 1,074 1,063 0 1,558 1,567 1,460 1,534 1,544 5 878 882 839 860 877

10 1,023 1,119 1,029 1,034 1,039 15 1,116 1,274 1,030 1,066 1,110 20 919 1,066 714 848 892 25 528 389 414 497 526 30 411 37 335 387 403 35 161 — 274 226 171 40 41 — 199 70 40 45 2 — 95 16 3 50 — — 31 1 — 55 — — 13 — — 60 — — 3 — — 65 — — — — — 70 — — — — — 75 — — — — —





–40 — — 5 4 2 –35 27 22 36 30 29 –30 54 64 66 57 53 –25 96 122 115 111 104 –20 179 209 182 175 179 –15 263 299 316 276 266 –10 578 628 641 620 598

–5 870 902 907 901 874 0 1,021 1,095 1,026 1,021 1,039 5 1,095 1,178 1,037 1,059 1,067

10 1,075 1,147 1,066 1,087 1,077 15 1,273 1,501 1,129 1,223 1,251 20 859 1,024 683 790 844 25 595 496 439 527 584 30 446 70 328 421 449 35 272 3 289 293 277 40 53 — 246 141 61 45 4 — 165 22 6 50 — — 70 2 — 55 — — 12 — — 60 — — 2 — — 65 — — — — — 70 — — — — — 75 — — — — —





–40 — — — — — –35 — — — — — –30 2 — 8 5 3 –25 55 45 102 66 56 –20 199 196 257 224 207 –15 335 397 336 350 336 –10 607 793 658 620 622

–5 1,096 1,201 1,085 1,098 1,094 0 1,217 1,177 1,178 1,206 1,232 5 1,034 985 990 1,036 1028

10 919 1,006 887 893 894 15 849 997 865 855 853 20 957 1,103 760 899 955 25 593 642 464 522 569 30 424 208 307 392 420 35 301 10 278 279 289 40 151 — 233 230 172 45 21 — 164 81 30 50 — — 130 4 — 55 — — 47 — — 60 — — 11 — — 65 — — — — — 70 — — — — — 75 — — — — —





–40 2 — 9 4 3 –35 24 22 27 25 23 –30 25 42 31 31 31 –25 87 85 132 92 86 –20 229 278 264 258 237 –15 331 412 348 330 334 –10 693 682 766 721 706

–5 1,162 1,271 1,129 1,168 1,167 0 1,133 1,126 1,063 1,105 1,135 5 846 921 886 849 832

10 1,157 1,239 1,204 1,199 1,152 15 1,127 1,192 917 1,045 1,121 20 775 996 640 736 759 25 460 472 359 412 456 30 420 46 302 370 433 35 264 — 240 280 255 40 49 — 223 148 54 45 — — 147 11 — 50 — — 79 — — 55 — — 18 — — 60 — — — — — 65 — — — — — 70 — — — — — 75 — — — — —

22





–40 — — — — — –35 — — — — — –30 4 1 10 6 5 –25 42 51 57 48 43 –20 138 165 184 156 146 –15 234 255 266 253 237 –10 533 616 603 551 537

–5 1,219 1,409 1,255 1,253 1,244 0 1,412 1,429 1,339 1,393 1,409 5 942 827 936 931 923

10 1,008 1,139 1,012 1,018 1,005 15 1,072 1,324 964 1,015 1,048 20 927 1,042 707 841 913 25 524 433 404 482 533 30 405 69 305 389 404 35 246 — 259 269 251 40 48 — 200 134 54 45 6 — 160 17 8 50 — — 79 4 — 55 — — 15 — — 60 — — 5 — — 65 — — — — — 70 — — — — — 75 — — — — —





–40 — — — — — –35 — — — — — –30 — — 3 — — –25 15 13 27 16 14 –20 63 74 112 79 74 –15 145 202 160 161 154 –10 336 357 445 385 370

–5 930 1,126 1,033 978 947 0 1,331 1,376 1,272 1,304 1,329 5 1,045 1,063 1,008 1,046 1,050

10 1,203 1,295 1,190 1,205 1,210 15 1,180 1,314 1,120 1,155 1,129 20 998 1,209 748 895 959 25 578 640 380 511 585 30 465 91 337 424 451 35 387 — 296 362 386 40 83 — 269 211 99 45 1 — 220 28 3 50 — — 107 — — 55 — — 28 — — 60 — — 5 — — 65 — — — — — 70 — — — — — 75 — — — — —





–40 — — — — — –35 3 2 8 5 4 –30 8 12 10 9 8 –25 20 37 50 23 22 –20 140 150 168 162 156 –15 311 339 408 336 312 –10 604 598 565 607 604

–5 990 1,176 1,080 1,021 1,008 0 1,171 1,184 1,111 1,152 1,168 5 1,065 1,137 1,050 1,048 1,079

10 1,043 1,067 950 1,006 1,019 15 1,072 1,218 1,023 1,069 1,068 20 874 1,111 731 815 846 25 584 547 420 528 575 30 443 174 323 413 440 35 295 8 266 333 296 40 126 — 239 163 140 45 11 — 199 62 15 50 — — 106 8 — 55 — — 42 — — 60 — — 11 — — 65 — — — — — 70 — — — — — 75 — — — — —

Appendix B—Exceedance Temperature Data for MS Structures The data in the following tables (Tables 28 to 35) show exceedance temperature data for field exposure structures in Starkville, Mississippi, from 1996 to 1999.

23

Table 28—Exceedance temperatures in MS dry, black-shingled structure, 1996




–40 — — — — — –35 — — — — — –30 — — — — — –25 — — — — — –20 5 3 17 12 7 –15 22 39 58 20 23 –10 99 153 148 118 100

–5 234 351 335 295 254 0 576 596 593 564 560 5 818 996 780 797 805

10 1,056 1,183 1,043 1,034 1,035 15 1,155 1,382 1,206 1,118 1,147 20 1,717 2,085 1,456 1,657 1,700 25 1,008 1,119 638 828 980 30 610 710 438 574 598 35 483 167 327 422 476 40 488 — 363 406 456 45 415 — 317 401 428 50 98 — 257 348 212 55 — — 315 188 3 60 — — 243 2 — 65 — — 202 — — 70 — — 48 — — 75 — — — — —





–40 — — — — — –35 — — — — — –30 — — — — — –25 — — — — — –20 — — 1 — — –15 13 5 33 19 14 –10 61 61 109 79 63

–5 182 214 314 228 196 0 640 733 728 668 625 5 983 1,147 929 952 970

10 1,095 1,327 1,081 1,069 1,096 15 1,223 1,543 1,144 1,170 1,200 20 1,500 1,854 1,348 1,439 1,466 25 1045 1,121 603 817 999 30 611 674 399 542 626 35 492 81 338 515 493 40 482 — 334 383 457 45 342 — 356 405 387 50 91 — 275 314 167 55 — — 287 157 1 60 — — 260 3 — 65 — — 184 — — 70 — — 37 — — 75 — — — — —





–40 — — — — — –35 — — — — — –30 — — — — — –25 — — — — — –20 — — — — — –15 — — 2 — — –10 20 7 86 34 22

–5 206 228 324 266 217 0 539 554 653 561 535 5 947 1,179 877 917 939

10 1,013 1,210 958 991 996 15 1,026 1,258 1,070 981 1,008 20 1,554 1,815 1,457 1,543 1,526 25 1,146 1,299 670 907 1,092 30 652 962 394 582 674 35 479 248 371 499 500 40 530 — 323 394 466 45 465 — 363 432 506 50 181 — 292 405 268 55 2 — 327 230 11 60 — — 315 18 — 65 — — 191 — — 70 — — 85 — — 75 — — 2 — —





–40 — — — — — –35 — — — — — –30 — — — — — –25 — — — — — –20 — — 12 4 — –15 17 11 21 20 16 –10 40 39 83 55 43

–5 185 177 352 242 191 0 453 482 463 453 454 5 778 999 840 796 765

10 1,147 1,373 1,088 1,097 1,101 15 1,194 1,510 1,215 1,161 1,199 20 1,611 1,944 1,332 1,487 1,558 25 1,079 1,234 651 911 1,074 30 684 784 453 624 688 35 522 207 383 493 503 40 527 — 372 413 485 45 370 — 317 432 457 50 148 — 311 358 196 55 5 — 309 186 30 60 — — 270 28 — 65 — — 198 — — 70 — — 84 — — 75 — — 6 — —

24

Table 32—Exceedance temperatures in MS wet, black-shingled structure, 1996




–40 — — — — — –35 — — — — — –30 — — — — — –25 — — — — — –20 — 3 12 3 — –15 20 39 18 21 22 –10 40 153 157 73 47

–5 286 351 358 306 290 0 647 596 596 620 613 5 860 996 810 810 811

10 1,122 1,183 1,012 1,031 1,072 15 1,180 1,382 1,176 1,120 1,150 20 1,780 2,085 1,493 1,684 1,740 25 923 1,119 681 901 996 30 568 710 479 618 609 35 542 167 331 435 507 40 453 — 363 452 470 45 350 — 303 404 387 50 13 — 304 290 70 55 — — 286 16 — 60 — — 250 — — 65 — — 146 — — 70 — — 9 — — 75 — — — — —

Table 33—Exceedance temperatures in MS wet black-shingled structure, 1997




–40 — — — — — –35 — — — — — –30 — — — — — –25 — — — — — –20 — — — — — –15 3 5 22 11 3 –10 29 61 106 43 33

–5 211 214 289 234 207 0 671 733 705 652 637 5 994 1,147 936 956 988

10 1,157 1,327 1,078 1,083 1,114 15 1,258 1,543 1,169 1,199 1,210 20 1,540 1,854 1,369 1,478 1,547 25 1,025 1,121 644 914 1,048 30 612 674 418 611 622 35 524 81 375 476 484 40 441 — 359 449 471 45 275 — 328 388 316 50 20 — 273 244 80 55 — — 309 22 — 60 — — 248 — — 65 — — 127 — — 70 — — 5 — — 75 — — — — —





–40 — — — — — –35 — — — — — –30 — — — — — –25 — — — — — –20 — — — — — –15 — — — — — –10 — 7 57 3 —

–5 221 228 313 248 212 0 543 554 624 551 519 5 973 1,179 906 917 939

10 1,070 1,210 942 1,014 1,038 15 1,064 1,258 1,061 983 1,063 20 1,697 1,815 1,484 1,588 1,580 25 1,085 1,299 722 1,009 1,150 30 644 962 437 673 680 35 550 248 392 471 537 40 546 — 371 482 549 45 346 — 316 494 388 50 21 — 330 297 105 55 — — 350 30 — 60 — — 258 — — 65 — — 171 — — 70 — — 26 — — 75 — — — — —





–40 — — — — — –35 — — — — — –30 — — — — — –25 — — — — — –20 — — 7 — — –15 8 11 17 13 10 –10 21 39 66 25 23

–5 196 177 305 227 218 0 452 482 477 450 466 5 831 999 832 786 811

10 1,169 1,373 1,095 1,102 1,091 15 1,349 1,510 1,232 1,219 1,252 20 1,728 1,944 1,355 1,573 1,661 25 1,077 1,234 698 1,021 1,100 30 677 784 490 683 697 35 610 207 415 511 569 40 479 — 369 502 540 45 162 — 349 434 265 50 1 — 341 193 57 55 — — 301 21 — 60 — — 245 — — 65 — — 135 — — 70 — — 31 — — 75 — — — — —

Documents

Roof Temperature Histories in Matched Attics in ... · roof temperature histories presented in this report provide reasonable estimates of actual thermal loads. In addition, the histories